The scientists demonstrate that benefits of animal models for humans are not proven and that this leads to funding being withheld from medically helpful research.

They describe how the National Centre for the Replacement, Refinement and Reduction of Animals in Research (NC3Rs) investigated animal studies conducted between 1999 and 2005 and observed flawed test execution that led to skewed results. In only 12 per cent of the studies were animals randomly allocated by the persons conducting the tests to a treatment or control group. Only 14 per cent of the persons responsible for evaluation were blinded with regard to allocating an animal to a treatment or control group. There was also evidence of distortions in the selective analysis and the result reports, and the significance of the results was exaggerated in the publications.

In the area of stroke research they cite Sutherland et al. 2012, who reported in a review that, despite the improved quality of preclinical research, there had been no increase in the applicability of test results during the previous ten years. According to the review, the concept of so-called Neuroprotection, with which scientists attempt to prevent reactivated neural tissue (penumbra) from dying using pharmacological or molecular biological methods, displayed promising results in animal tests but did not lead to success in clinical practice. They cited a number of reasons for this, such as heterogeneity in the origin of strokes in humans and a lack of methodical correlation between preclinical and clinical studies. In addition to evolutionary species differences, they also addressed physiological and epigenetic differences, especially between phylogenetically closely related species.

Even if the quality of preclinical research and reporting improves, the authors believe that would not change much, as the ability to predict human reactions to a given treatment based on the results of animal tests remains restricted by species differences in molecular biology and metabolic paths.

However, the scientists come to the conclusion that therefore it is preferable to invest more funds in clinical research and less in basic research.

In my opinion this does not make much sense, as for safety reasons preclinical research always precedes clinical research. Perhaps it would be better to invest much more in new methods as alternatives to animal experiments.

Sources:
Pound, P, Bracken MB & Dwight Bliss, S (2014): Is animal research sufficiently evidence based to be a cornerstone of biomedical research? Analysis. BMJ 2014;348:g3387.

Godlee, F (2014): How predictive and productive is animal research? BMJ 2014;348:g3719 doi: 10.1136/bmj.g3719

Sutherland, BA, Minnerup, J, Balami, JS, Arba, F, Buchan, AM & Kleinschnitz, C (2012): Neuroprotection for ischaemic stroke: Translation from the bench to the bedside. International Journal of Stroke 7: 407–418.